Wavelet analysis of SAECG to identify patients with conduction defects at risk for sudden cardiac death.

The aim of this study was to determine how Wavelet transform analysis of signal-averaged ECGs can identify patients with conduction defects who are at high risk for development of ventricular tachycardia. In this study, 34 SA-ECGs and programmed electrical stimulation (PES) reports were obtained from the OSU Department of Cardiology Database (1988-1996) and divided into two groups: 17 patients that had inducible monomorphic VT by PES (VT+) and 17 that showed no arrhythmias (VT-). We used Morlet's wavelet to analyze the X, Y, Z, and RMS vector magnitudes in each group. The mean duration from the peak of the RMS vector magnitude to the QRS offset was statistically different with a T value (2-tailed distribution, unequal variance) of 0.033. We noted statistically significant (p < 0.0001) differences in Wavelet energies for 44 msec after the peak of the RMS vector magnitude largest in the Z lead, the first 22 msec, and frequency bins less than 131 Hz. Although no clinical marker could be determined using Wavelet analysis to distinguish the the VT+ from the VT- group, the results from this study show that their SA-ECGs are indeed different even though the optimal analysis has not yet been devised.